Everything you know about conversion optimization is not true.

Friends, we want to share with you an adapted translation of a revolutionary article by Larry Kim in the blog of the company WordStream . The article will turn your idea of ​​conversion, testing of landing pages and what you need to do to increase efficiency.
The article presents tips for improving site conversion with examples from the Witget project , a site conversion enhancement service.


Conversion is a key indicator of the effectiveness of contextual advertising: if you do not convert traffic from contextual advertising into buyers, then why bother advertising at all? Conversion optimization allows you to convince more of your potential buyers to take a targeted action.

But what level of conversion is high?
')
If you have already achieved a conversion of 3%, 5%, or even 10%, is this the level that suits you?

We analyzed the sites of thousands of AdWords customers with total annual expenditures of $ 3 billion and found that for some advertisers, site conversion is twice or three times the average. Do you want to be at the average level or are interested in your site demonstrating the best results, increasing exponentially in comparison with others in the same industry?

The analysis of a large amount of information about landings and conversions allowed us to identify some common characteristics of the landing pages with high conversion. What do they have, what don't you have? Believe it or not, there is no border between you and a conversion two or three times higher than what you see today. However, the way in which such results can be achieved is completely opposite to the generally accepted notion of conversion optimization.

This article describes step by step the conversion process, confirmed by statistical findings from the best (and worst) existing advertisers on the market.

Today we will cover the following topics:

- Why are generally accepted notions of conversion optimization unjustified?
- What is high conversion?
- How to achieve high results?

Ready to find out why everything you know about conversion optimization is wrong? So we start ...

Why are generally accepted notions of conversion optimization unjustified?

Finding out that your esteemed experts turned out to be wrong in the same way as if the child first learned that Santa Claus is not real. Under an elegant suit the usual unshaven guy actually hides. Everything that you learned about conversion optimization is just that: sparkling and attractive outside, but with serious flaws inside.
How can everyone perceive conversion optimization so wrong? It's obvious: if you sing the same song that everyone sings around, you really can never achieve results above average. When all the gurus preach the same optimization principles that all your competitors listen to, how can you be different from them?

The classic conversion optimization tricks don't really work.

Ladies and gentlemen, your attention is invited to the tale of the Great Optimization of Conversion. One day, a homegrown marketing guru told how important site optimization is. He cited as an example such developer actions as changing the key color or font size or placing a picture, and as a result ... Attention! Conversion of the advertiser "rapidly" increased by 2-7%.


Amazing isn't it ?! No, not really surprising. These are generally accepted basic recommended A / B testing methods. Such optimization should be performed regularly, and perhaps as a result there will be a slight increase in conversion - but this is unlikely to increase conversion by 10% or more.

Let me show you what happened to the achievements that were made possible thanks to these little changes on your page.

We offer an example of split testing for landing; the bottom gray line is the first version of the page with which we work. The blue line is the second option that we optimize. First, the new page is much higher than the old one in terms of performance. Great, isn't it?


Except that progress does not last long. Indeed, the “best” page eventually ends up showing growth. We started testing for a long time and at the same time and saw this trend. We call this the dilemma of prematurely stopping testing. You see a potential customer expected in the near future, but it soon disappears.

This, of course, is not always relevant. However, we found out that in most cases small adjustments, such as changes in the spacing between lines, font colors, and others, are equal to small achievements. If you want to achieve a fast, serious and long-lasting increase in conversion, then you need to leave in the past these “crutches”, which contribute to an increase in conversion only for a couple of days or weeks.

Why it happens? Often because you look at small amounts of data that are not representative. If you're considering 50, 100, or even 200 conversions as a result of the entire test, small changes may seem more effective than they really are. A pair of conversions can mean an increase in conversion of 4% if the total number of conversions is 50. This is due to the fact that your sample is not large enough to start with.

It's time to stop moving chairs from place to place.

When optimizing a landing page, it may turn out that you are actually busy with performing minor actions with low efficiency, as if you were moving chairs from place to place on the Titanic deck. We need to leave in the past this approach, replace it with effective tactics and optimization actions that will significantly change the results and revenues.


The first thing we need to know is:

What is a good conversion?

Hint: it is much higher than you might think.

According to conventional wisdom, a good conversion is somewhere between 2% and 5%. If you are at 2%, an increase to 4% looks like a significant jump up. You doubled the conversion! Well, well, congratulations. However, you again stopped at the level of averages.

In this study, we started with all the accounts that could be analyzed, and returned to the period of the past 3 months. We have excluded those accounts for which:
- conversion tracking was not properly installed,
- there were low conversion volumes (<10 conversions per month),
- low intensity of visiting accounts (<100 clicks per month).

We left hundreds of accounts for our analysis. Then we’ve mapped the accounts that matched a specific conversion rate.


So what is a good conversion? About 1/4 of all sites have less than 1% conversion. The average was 2, 35%, and 25% of the best sites have twice as much - 5, 31% - or more. Look at the extreme red column on the right - the first 10% of AdWords advertisers have a website conversion of 11, 45%.

Remember, these figures do not belong to individual landing pages - these advertisers perform a conversion of 11, 45% and higher throughout the site.
Obviously, this is not an exceptional situation - it is quite feasible. If the conversion is currently 5%, then you are ahead of 75% of advertisers ... but you still have great potential for growth!


You should increase the rate to 10%, 20% or even higher, increasing the conversion 3-5 times higher than the average.

But in my industry, the conversion is lower ...

This situation is quite possible. We systematize conversion data by industry. And this is what we found when analyzing the four main industries:


There is a difference in the conversion by industry. E-commerce is noticeably losing in terms of average conversion, especially compared to finance. But do not compare by industry. Look better at conversions of the top 10% in your industry and you will see a lot of potential for growth.

Of course, there is a downside: if you work in an industry with high rates, such as finance, then 5% is really not a fantastic conversion. Therefore, to look at the average in the hospital is wrong. For finance, 2-3% is just a shame.

In general, the top 10% show indicators five times higher. That's where you need to grow!

Even if the average conversion rates are lower in your industry, the best advertisers are 3-5 times or more ahead of you.

Top 10% of landing pages: what makes them work?

What do the top 10% of landing pages look like - and how do they win the competition? We reviewed 1,000 landing pages and performed a qualitative analysis to find patterns.

I offer you the five best tips that will help you achieve the results of a better landing page:

1. Change offer

Considering all the high performance landing pages, we see amazing creative and individualized offers.

Often, companies offer for free what money is for competitors. Lawyers, for example, offer a free initial consultation. Software companies will offer free trial versions. Such offers are ordinary and not creative.

How to make an offer unique?
Our client in the designer Witget made a vitget - a pop-up window with the contact collection form, where he offered to receive an e-mail dictionary intended for potential clients. Given that new customers come to the site a day to an average of 300 people, 7 to 10% of them leave a request for a dictionary a day. Further work with potential clients is carried out through e-mail marketing.


How to find out that the offer has become ineffective? If the conversion stopped at 2% or lower, all is not hopeless. To find out how bad the offer is, you need to ask the customers about it. For example, you can place a special dynamic widget that is located at the bottom of the page and offers help.
Witget in expanded form:


This witthall allowed the client - the company that is involved in the implementation of CRM-systems - to understand what the site visitors actually lack before the target action is performed. It turned out that the primary information is collected by the office manager, who needs to provide management with concise information about what the system is and how it is better than others. As a result, the company prepared a manual and competitive comparison in the form of pdf and made it possible to download it on any page of the site. Office managers saved time on the compilation of comparative tablets and sent out a manual to management. Since the manual represented the company in a favorable light, the number of appointments to demonstrate the system increased by 23%.

Brainstorm, ask your customers, and you will have more unique offers that you can try to implement. You will never know which one is better until you try new ones.

2. Change the sequence of actions

Sometimes you set limits to increase conversions, without even realizing it.


In the above example, the first version of the landing page is presented, which demonstrates how much information people are forced to learn before downloading their version of the software for trial use. Obviously, this is too difficult for most customers. Such a waste of time is disappointing, it is unlikely that the user will be interested in such a landing page.


Here is the result of the client's work with the new version of the landing, which, as it turned out, is much more efficient than the previous one. They changed the sequence of actions so that any person could download and install the file. At the last stage, the user is asked to register the software. By this time, they have spent 10 or 15 minutes on the software and are more likely to spend time filling out the information form.

The changes were so effective that the page received a large number of conversions. Changing the sequence of actions helped increase conversions, and also allowed managing the quality of the leads in a much more efficient way.

We offer another good example in which the site owner works with three categories of potential customers at once: those who are in search, those who want to help relatives, those who need urgent intervention.


This is extremely effective not only for conversion, but also for segmentation during remarketing and attempting to process leads.
So what is the main idea here? Do not be afraid to contact different groups of your customers with different offers.

3. Use remarketing as a tool to optimize conversion.
On average, 96% of people who visit a website will not be potential buyers or will not make a sale. Remarketing helps you deliver targeted, relevant messages to different people on different resources: in the mail, in videos on YouTube, on social networks, in search engines, on various subject sites.

4. Work out 10 landings to find 1 effective

If you want to achieve the same indicators that are observed in these top 10% of landing pages, you need to repeat the above steps several times and perform testing regularly.

On average, you should test four unique landing pages - making changes to offers, the sequence of actions and sending messages. If you want to find the best page that will help your conversion to be 3-5 times higher than the average - you need to test at least ten landing pages.

Here we analyzed the account in the e-commerce industry with 1000 unique landings. About a third of the traffic is spent on the best landing pages. If we dig deeper, we will see that about 80% of the traffic goes to only 10% of landing pages.


You do not need to create thousands and thousands of landing pages. You need to find the best of those that you already have, and focus your efforts on them. How can you prove their effectiveness? Discard too much, stop wasting time on pages with low rates - just get rid of them. If you have only one excellent landing page, it would be better to focus your efforts on it.

I offer one more proof that working at night, creating hundreds of options for landing pages, is not the best use of your time.


The bottom line is that we compared thousands of sites with unique landing pages in terms of conversion. We do not see a strong correlation between the increasing number of landing pages and increasing conversion.
If you are targeting the best pages, quantity does not necessarily mean quality.

5. Click on conversion rate.

This title is designed to sober you up. You do not need conversion for conversion. If you offer a million dollars in the offer and 100% of people fill your form, can you rejoice in it? The question is not how much the form was filled in, the question is how much did you buy later! Soberly look at the indicators and think first of all about profit, and secondly, how to achieve this profit.

Conclusion

So, what have you learned from all of the above? I hope you will be consistent in using the following recommendations to optimize the conversion:
1. A large number of ways to optimize the landing is similar to moving from place to place of chairs on the deck of the Titanic. Small changes = small results.
2. Extremely focused and strategic optimization of landing pages will lead to a 3-5-fold increase in conversion and improve the quality of leads.
3. In some industries, even 5% conversion is not impressive. If your conversion is delayed at the level of 2-5%, you have something to work on.
4. Offer creative offers and constantly test them to find the one that is most effective for your audience. But do not stop the development. Offer always something new.
5. Identify the obstacles that prevent potential buyers from taking action, and eliminate this interference from their path by changing the sequence of actions. Test various options to find out exactly which conversion path works best for your audience.
6. Use remarketing to get the attention of people who have left the site without performing a targeted action.
7. Perform quality testing. You need to test 10 options for a unique landing page to find the best one, but testing should go beyond the bounds of a primitive font color change.
8. Remove excess on the site and exclude pages with the lowest rates. Focus on 10% of the best landing pages that give 80% of leads.
9. Always, always keep an eye on your goal, which is to achieve high sales or generate more leads, which are most likely converted into sales. Do not allow high conversions to take precedence over lead quality, otherwise you will lose leads that are more likely to be converted into purchases. You need to find a middle ground when everything works like a well-oiled mechanism.

Editor - Anna Chaschina .

Thank you for your attention and for your positive feedback. , , YouTube- , Witget.com.